Loot Table Expected Value Calculator

How this is calculated

EXPECTED VALUE IS CHANCE TIMES VALUE, ADDED UP. A 2% drop worth 5,000 contributes 100 to every run whether or not it appears - that is what expected value means. It is the correct way to compare two farms, and it is also the figure most likely to be misread, which is what the concentration output below is for. ⚠️ THE AVERAGE IS NOT WHAT A SHORT SESSION LOOKS LIKE. When most of a loot table's value sits in one rare item, the expected value describes a long run and almost nobody's evening. At the defaults, 30% of the value comes from a drop appearing once in 200 runs - so a typical session earns well BELOW the average, and the average is only recovered by the occasional session that goes far above it. That is not a flaw in the arithmetic; it is what a skewed distribution does, and knowing which kind of farm you are on is more useful than the headline number. THE CONCENTRATION FIGURE IS THE ONE TO READ SECOND. Low, and the farm pays steadily and predictably. High, and you are effectively buying lottery tickets with a good expected price - fine if you can farm it for long enough, frustrating if you cannot. ⚠️ VALUES ARE YOURS AND THEY ARE THE SOFT PART. Market prices move, and an item you need is worth more to you than its sale price while an item you already have is worth its market value at best. There is no objective figure here, which is why the tool asks rather than assuming - and it is why comparing two farms is more reliable than trusting one farm's absolute number, since your valuation errors tend to cancel. DROPS ARE ASSUMED INDEPENDENT AND SINGLE. Real loot tables often are not: some roll one item from a weighted table so the chances are mutually exclusive rather than independent, some guarantee a minimum, some drop stacks rather than single items, and many apply pity or bad-luck protection. Expected value survives most of that - it adds up regardless of how the rolls interact - but the any-drop and runs-to-see figures assume independence and will be wrong where the table does not work that way. RUN TIME SHOULD INCLUDE EVERYTHING. Travel, setup, repairs, restocking and the walk back all cost real time, and a farm with a fast fight and a slow reset is worse than its combat time suggests. Timing a few full cycles beats estimating. THIS COMPARES FARMS; IT DOES NOT VALUE YOUR TIME. Whether an hourly rate is worth doing is a separate judgement, and one worth making.

Common questions

What does expected value actually mean here?

Each drop's chance multiplied by its value, all added together. A 2% drop worth 5,000 contributes 100 to every single run - not because you get 100 each time, but because over enough runs that is what it averages to. Adding those contributions gives the value of one run, and multiplying by runs per hour gives the rate. It is the correct basis for comparing two farms, because it accounts for both how often something drops and how much it is worth, which neither figure does alone.

Why does my actual loot never match the average?

Because when most of the value sits in one rare item, the average describes a long grind rather than an evening. At the defaults, 30% of the expected value comes from a drop that appears once in 200 runs - so a typical session earns well below the average, and the average is only recovered by the occasional session that goes far above it. That is not a fault in the arithmetic, it is what a skewed distribution does. The concentration output tells you which kind of farm you are looking at before you commit an evening to it.

How should I read the concentration figure?

As a measure of how lumpy the payout is. Low concentration means the value is spread across common drops, so the farm pays steadily and a short session earns roughly its share - good if you play in small blocks. High concentration means you are effectively buying lottery tickets at a reasonable price, which is fine if you can farm long enough for the average to assert itself and frustrating if you cannot. Two farms with identical expected value per hour can be completely different experiences, and this is the number that tells them apart.

What should I put for the values?

Whatever the item is worth to you, which is the softest part of the calculation and worth being honest about. Market prices move, an item you actually need is worth more than its sale price, and an item you already have is worth its market value at best - possibly less if selling it is slow. There is no objective figure, which is why the tool asks rather than assuming one. The practical consequence is that comparing two farms is far more reliable than trusting a single farm's absolute number, because your valuation errors largely cancel out between them.

Does it handle loot tables that roll one item?

Expected value, yes. The other figures, no. If a table picks a single item from a weighted list, the chances are mutually exclusive rather than independent - but expected value still adds up correctly, because each item still contributes its chance times its value regardless of how the roll works. What breaks is the any-drop figure, which multiplies independent probabilities and will understate a table that always drops something. The runs-to-see figure is also affected by pity systems and bad-luck protection, which many games apply and none of which this models.

What should the run time include?

Everything, not just the fight. Travel to the location, setting up, the encounter itself, repairs, restocking consumables and the walk back all cost real time, and a farm with a fast kill and a slow reset is considerably worse than its combat time suggests. Timing a few complete cycles with a stopwatch beats estimating, because the parts people forget - loading, banking, the run back - are exactly the parts that differ most between farms and decide which one actually pays better.

Does a higher expected value always mean a better farm?

Only if the value figures are right and the session length suits the concentration. Beyond that there are things this deliberately does not weigh: whether you enjoy it, whether it teaches you anything, whether the drops are things you want or things you sell, and whether the hourly rate is worth your time at all. That last one is a genuine judgement rather than a calculation - working out that a farm pays a certain rate an hour is useful precisely because it lets you decide whether that rate is worth doing, and sometimes the honest answer is no.

Take it further with AI

Copy this into ChatGPT or Claude with your own numbers filled in. It hands over the figures this calculator worked out, so the answer is built on real arithmetic instead of a guess.

I used the Loot Table Expected Value Calculator at https://www.bfcbrilliance.com/tools/loot-table-expected-value-calculator.

What I entered:
- Minutes per run (min): ___
- Common drop - chance (%): ___
- Common drop - value (value): ___
- Uncommon drop - chance (%): ___
- Uncommon drop - value (value): ___
- Rare drop - chance (%): ___
- Rare drop - value (value): ___
- Rarest drop - chance (%): ___
- Rarest drop - value (value): ___

What it calculated:
- Expected value an hour: ___
- Expected value of one run: ___
- Runs an hour: ___
- Share of the value in the rarest drop: ___
- Runs to expect one rarest drop: ___

Use those figures as given — they are already worked out, so please don't recalculate or estimate your own. Help me turn them into a plan: what to buy or do, in what order, roughly what it should cost, and the mistakes people most often make with this job.

More drop rates & grinding tools